The present invention relates to a semiconductor manufacturing apparatus, its diagnosis apparatus and operating system which are arranged to maintain the semiconductor manufacturing apparatus, and a method of manufacturing a semiconductor.
The semiconductor manufacturing apparatus has traditionally given an alarm to an operator of the apparatus with beeps on communication arrangements, a section for managing the apparatus itself or the like if the apparatus suffers from an error or an unfavorable condition. Moreover, the occurrence of an abnormal condition like an error is reported to a manufacturer or a supplier of the apparatus, an enterprise of diagnosing or operating the apparatus, or an enterprise of supplying such services. In response to the alarm, the operator or engineer of the apparatus searches the information on the abnormal condition reported thereto, checks for the apparatus, and repairs it if possible.
In place, if the abnormal condition (often referred to as an abnormality) concerns with a quite significant portion or a portion where the highly technical knowledge is necessary to the check, the skilled person belonging to the foregoing enterprise or special section comes to the place where the apparatus is installed for checking therefor. Then, the skilled person reports the result to the responsible section of the manufacturer or diagnosing or operating the apparatus. In response, the skilled person repairs the apparatus according to the process and the measures determined on the result reported therefrom.
However, the places where the semiconductor manufacturing apparatus are scattered anywhere domestic or worldwide. Under this circumstance, in some cases, a considerably long distance between the places requires the skilled persons belonging to the diagnosis or operating section to take so long a time in moving from one place to another that the quick response to the alarm cannot be substantially supplied to the buyers or users of the apparatus. Also, a considerably long distance between the buyer or user of the apparatus, the section of supplying a service like a maintenance, and the manufacturing place brings about a difficulty in quickly responding to the alarm reported from the place where the semiconductor manufacturing apparatus is installed.
Further, in the apparatus installed place, these skilled persons have difficulty in properly responding to the abnormality, because when they come to the place, at least the time taken in moving them from their place to the apparatus installed place is passed so that the state of the apparatus cannot be precisely determined. Further, the skilled persons also have difficulty in reporting the detailed state of the apparatus to the section in charge of designing or manufacturing the apparatus through the communication arrangement. The rough, that is, often inaccurate information on the state of the apparatus does not easily allow the approximate response to the abnormality to be provided to the skilled persons located nearby the apparatus.
In order to overcome these difficulties, the concerned technology has been devised wherein the information on an abnormality like an error, a failure or a trouble occurring in the apparatus is reported from the apparatus side through a communication arrangement located therebetween and the proper response to the abnormality is reported back to the apparatus side. This type of technology is disclosed in JP-A-7-282146 (First Prior Art) or JP-A-10-21079 (Second Prior Art). These prior arts are arranged to report the information of an abnormality like an error and to report back the measures determined on the information through a network like a LAN (Local Area Network) served as the communicating arrangement.
The first prior art (JP-A-7-282146) is arranged to detect an operating timing of each unit inside the apparatus through the use of a detector like a wafer sensor or a pitch sensor, determine an abnormality occurs if the detected timing is out of the allowable range, and then report an alarm to a computer for managing the apparatus. Further, the computer is connected to a workstation through the communicating arrangement (network), the workstation having an expert system operated thereon. In response to an alarm including the number and the abnormality code of an I/O where the abnormality is detected, the expert system operates to transmit the specific method of responding to the malfunctional portion to the computer of managing the apparatus as the diagnosed result and then display the pairs of the failure candidates and the responding methods thereto on the display unit.
Further, the second prior art (JP-A-10-21079) concerns with the failure diagnosing apparatus in which if an abnormal state occurs in a semiconductor manufacturing apparatus, an alarm code having a predetermined data format is reported to a computer in charge of diagnosing the semiconductor manufacturing apparatus through a communication network. This computer stores the rules of inferring the causes of the abnormality and the responding methods thereto from the reported alarm codes. On these rules, the computer infers the cause of the reported abnormality and its measures from the alarm code. Then, the display provided in the diagnosing apparatus operates to display the corresponding message to the diagnosed result.
The foregoing prior arts are both arranged to connect the semiconductor manufacturing apparatus with a computer for diagnosing this apparatus for an abnormality through the communicating arrangement (LAN) and report the abnormality through the communicating arrangement if it occurs. Then, this computer operates to analyze the cause of the abnormality and its concerned portion, operate the measures for the abnormality, and selectively outputting the measures. This diagnosed result is transmitted to the display unit or the semiconductor manufacturing apparatus on the side of which it is displayed on the display unit.
However, the foregoing prior arts are arranged so that the computer may compute the state and the cause of the abnormality. Hence, the diagnosis of the abnormality is executed on the predetermined rules stored in the computer. It means that these prior arts do not take into consideration the provision of the information on the apparatus so that the skilled personnel in charge of the concerned apparatus may easily grasp and determine the state of the apparatus and the cause of the abnormality.
In order to execute the foregoing diagnosis through the computer, the storage unit located in the computer is required to systematically store the knowledge on the relation between the symptom and the cause of the abnormality so that the knowledge may be reused at any time. The obtaining and the systematic storage of the knowledge are required to accumulate the experiences and the knowledge of the skilled engineers and the service personnels. Moreover, the knowledge required to determine each cause of the abnormality according to the specification of the operation and the procedure are also requested to be accumulated as the semiconductor manufacturing apparatus is being promoted. The accumulative obtaining of such a knowledge normally needs a long time. In order to obtain such a knowledge, it is necessary to supply means of providing the information on the apparatus so that the personnels in charge of checking the apparatus may obtain the precise knowledge on the operation and the process of each type of semiconductor manufacturing apparatus. Further, if the abnormality is detected, the apparatus is stopped or its operation is restricted. This condition needs to improve the working efficiency of the apparatus. This thus requests to obtain the knowledge required to determine the cause of the abnormality for a short time. These are not considered in the foregoing prior arts.
In general, the semiconductor manufacturing apparatus needs a plurality of operations of the apparatus for processing the raw materials. It means that there may be other causes than the operation being executed when the abnormality occurs is a direct cause. Specifically, in the semiconductor manufacturing apparatus constructed to process the raw materials through the use of a plurality of gases and working materials, the cause on which the abnormality occurs in the apparatus may be often ascribed to the mutual relation of the operations being executed in the apparatus. In such a semiconductor manufacturing apparatus, it is preferable that the informations of not only a single operation but also the relation of plural operations may be used for determining the cause of the abnormality. In order to precisely the cause of the abnormality, it is important to estimate how significantly each operation of the apparatus has an effect on the apparatus. Likewise, if a plurality of operations being serially executed have an effect on the abnormality, the sensor may often sense the abnormality of only a single operation being executed for a short time and display it. In this case, it is difficult to precisely determine the cause of the abnormality.
In consideration of the aforementioned respects, the foregoing prior arts do not take into consideration means of providing information so that the user may easily determine the state of the apparatus and the cause of the abnormality. Hence, the user cannot precisely determine the cause and the state of the abnormality taking place in the apparatus, which leads to increasing the time when the apparatus is stopped and repaired and thereby lowering the operativity of the apparatus and increasing the manufacturing cost of the semiconductors.
Further, in order to more precisely determine the cause of the abnormality taking place in the semiconductor manufacturing apparatus or determine the state of the apparatus where the abnormality occurs, it is preferable to check the apparatus as the apparatus is actually operated in the condition where the abnormality occurs. However, in a case that the apparatus is stopped in response to the report on the occurrence of the abnormality, it is difficult to operate the apparatus again. Even if it is possible, it is difficult to reproduce the same condition as the condition where the reported abnormality occurs. This difficulty makes it impossible to precisely check the operation of the apparatus, which disadvantageously results in impairing the precise determination of the abnormality cause.
In a case that the section in charge of maintaining and repairing the apparatus is located remotely from the place where the apparatus is installed, while the personnels in charge of checking the apparatus is moving to the place, the apparatus cannot be worked normally. If the section or the personnels in charge of maintaining or repairing the apparatus is located remotely, the cause of the abnormality of the apparatus cannot be properly determined for a short time, which leads to lowering the working efficiency of this apparatus, thereby disadvantageously increasing the manufacturing cost of the semiconductor.
As described above, no consideration is taken into the means of allowing the user to precisely grasp and specify the state of the abnormality-caused apparatus and the cause of the abnormality for a short time.
The foregoing prior arts thus need a long time until the cause of the abnormality is specified, so that the apparatus cannot be efficiently promoted. For shortening the time, it is possible to take the measures therefor such as location of the section in charge closer to the place where the apparatus is installed and distribution of the personnels in charge of maintenance and repair to each apparatus. However, these measures make the cost of the personnels in charge increase and thereby the operating efficiency of the apparatus inferior. It means that the foregoing prior arts do not sufficiently consider the means or system required to efficiently promote the semiconductor manufacturing apparatus where the abnormality occurs. In particular, no consideration is taken into the efficient promotion of a group of these apparatuses in a place like a factory where a plurality of apparatuses are installed.